Undertaking innovative cancer research can require input from teams of scientists with a mixture of backgrounds, including molecular biology, oncology, medicine, epidemiology, biostatistics, genomics/genetics, and computer science. Researchers with interdisciplinary training across these fields are extremely valuable to such teams, as they can act as conduits for the integrated work necessary to accomplish some of the most promising and forward-looking cancer research. Due to the exclusive nature of training within these fields, however, there are limited opportunities for investigators to obtain the knowledge that bridges these disciplines. To help remedy this problem, we propose here the continuation of this R25T program to provide postdoctoral training in the computational genetic epidemiology of cancer. This program defines a novel, transdisciplinary area of training at the intersection of cancer research, epidemiology, biostatistics, genetics, and computer science. The program's structure is defined by three key requirements. First, all trainees will take a specialized core curriculum of five courses that cover the individual disciplines as well as their intersections. Second, the trainees will undertake additional didactic experiences selected to complement their educational and research background. Third, all trainees will obtain research experience by collaborating with multiple mentors on high-level computational genetic epidemiology of cancer projects. As an extension of this research experience, each trainee will be required to write and defend a mock NIH proposal. Cancer researchers obtaining training in this program will have the skills vital to deciphering the complex pathways comprising genetic and environmental risk factors for disease. In doing so, they will ultimately be able to provide clinicians and their patients with invaluable information for the prevention and treatment of cancer. [unreadable] [unreadable] [unreadable]